A novel radio source search algorithm using force field vectors and received signal strengths

The indoor radio source search using received signal strength (RSS) is difficult due to the multipath effects. Robots driven by the RSS gradient-based searching methods are always stuck by local basins which are the areas around local maxima. In this paper, we propose a force field searching (FFS) algorithm which is cost efficient in environments with RSS local basins. The FFS algorithm fuses the geometry layout information with RSS. Virtual attraction forces weighted by RSS gradients are modelled to guide the robot to the radio source. Furthermore, the travel costs of FFS and the gradient ascent with correlated random walks (GACRW) algorithm are derived as functions of the searching space size, the local basin size and the local basin number in gradient fields. The analytical expressions are confirmed by simulations. It is shown that FFS is much faster than other algorithms in simulations driven by real data.